The present invention relates to the field of data collection and, more particularly, to collecting interest data from conversations conducted on a mobile device to augment a user profile.
The collection of user data is a common practice among Web-based applications and services. Often, a user is required to fill out a questionnaire, typically called a user profile, when registering for such applications and services. For example, an online social network community requires a new member to list a variety of personal information that allows other members to find people with common interests.
As noted above, user profiles are created and modified manually by a user. While effective, in a simplistic sense, this process is fundamentally flawed by completely relying upon the user to input data. In particular, this process relies upon a user's self-awareness and ability to express their interests. For example, a user who walks home from work through a park everyday may not consider “nature walks” as an interest because the user associates walking in the park with work and not enjoyment.
Further, a user is often unaware of new and/or casual interests, especially when there is a significant amount of time between occurrences or resources required. For example, a fledgling interest in jazz may go unnoted if the user must plan expensive trips to other locations to attend performances. Additionally, a casual interest can go unnoticed when a user participates only with a specific group of people, such as playing pool with coworkers occasionally after work.
When a user identifies a new interest, the addition of this information to the user profile must be made manually by the user. Updating one's user profile is often not a high priority for many users. Thus, user profiles tend to capture a recent, but static listing of a user's interests.
When participating in a new and/or existing interest, a user typically engages in a variety of conversations related to the interest. These conversations, which are often conducted on mobile devices, are an untapped data resource that can assist in the identification of a user's interests. Conventional systems for capturing interest data for a user profile ignore this wealth of data.